Análise dos fatores de risco para peritonite bacteriana espontânea em pacientes cirróticos e do perfil da flora infectante com o uso de antibióticos profiláticos / Analysis of risk factors for spontaneous bacterial peritonitis in cirrhotic patients and the ascitic fluid microbiology with use of prophylactic antibiotics

Valdinélia Bomfim Barban Sposeto

INTRODUCTION: Invasive procedures and the decline of the liver function have been considered predisposing factors for spontaneous bacterial peritonitis (SBP) in cirrhotic patients. In spite of the predominance of gram negative, the incidence of gram positive agents is increasing in literature. OBJETIVES: To analyze the risk factors for SBP in cirrhotic patients and to assess if there is increase in the frequency of infection by gram positive agents, according to the use of antibiotics. METHODS: In this retrospective study, the results of 1.114 paracentesis carried out in 348 patients from 2005 to 2007 in the Department of Gastroenterology of the University of São Paulo were enrolled. According to the result of ascitic fluid leucometry, two groups were formed: with and without SBP. The following factors were assessed: aspartate aminotransferase; alanine aminotransferase; bilirubin; INR; creatinine; use of propranolol and hemodynamic response; previous gastrointestinal hemorrhage; hypovolemic shock; endoscopic therapy of esophageal varices; vesical catheter, indwelling vascular catheter, severity of the underlying liver disease (scores Child-Pugh, MELD and MELD-Na); concurrent bacterial infections and the frequency of gram positive bacteria according to the use of antibiotics. RESULTS: 852 paracentesis performed in 303 patients were included. The most prevalent etiology of cirrhosis was hepatitis C virus infection (25.4%), followed by alcoholic (24.1%). The diagnosis of SBP was established in 82 (9.6%) paracentesis, 27 (33%) of them were classical SBP and 55 (67%) were negative-culture SBP. In the SBP group, we found higher levels of bilirubin and more enlarged INR (p<0.0001 e p= 0.0016, respectively). There was no difference between the groups regarding the risk of SBP and the use of betablockers (32.9% versus 37.3%, p=0.533) or hemodynamic response to propranolol therapy (68.2% versus 70%, p=1.00). The following parameters did not reach statistical significance: gastrointestinal bleeding (6.1% versus 2.5%, p=0.074), endoscopic sclerotherapy of varices (2.4% versus 0.8%, p=0.78), vesical catheters (4.9% versus 2.3%, p=0.138), vascular catheters (2.4% versus 1.7%, p= 0.649). The SBP group had a higher frequency of Child C status patients, 51% versus 37%, (p=0.022) and higher frequency of hypovolemic shock 2.5% versus 0.3% (p=0.0484). There was no difference in the frequency of SBP in patients with or without concurrent bacterial infections (p=1,00). In the SBP group, gram positive staining bacteria were found in 55.6% and gram negative in 44.4% (p=0.3848). We found no relationship between gram positive bacteria infection and the use of quinolones (p=1.00). Every single point increased in the MELD score increased the risk of SBP in 1.059 times [95% IC: 1.0266; 1.0930] or by 6%. There was no significant difference in the odds ratio for SBP according to the stratification of MELD values. Every single point increased in the MELD-Na increased the risk of infection in 1.0283 times [95% IC: 1.0073; 1.0497] or 2.8%. Nevertheless, every point increased in the MELDNa between 6 and 15 increased the probability of infection in 1.3371 times [95%] IC: 1.0230; 1.7476], between 16 and 24 in 3.2371 times [95% IC: 0.1958; 53.5291] and higher than 24 points in 14.2663 times [95% IC: 1.2441; 163.5990]. CONCLUSIONS: Patients with SBP had higher levels of bilirubin and INR, higher frequency of hypovolemic shock and more severe underlying liver cirrhosis, as assessed by the Child-Pugh score, MELD and MELD-Na, indicating that the decline of the liver function is the main risk factor for developing SBP in cirrhosis. The use of betablockers and the hemodynamic response to propranolol were not associated to protection against developing SBP. The odds ratios for developing SBP increased according to the stratification of MELD-Na values, but not according to MELD stratification. There was no significant difference in the frequency of gram positive and gram negative infections in patients with SBP. The use of quinolones was not associated with increased frequency of gram positive infections in this series .

AUTOR(ES)

DATA DE PUBLICAÇÃO

RESUMO

ASSUNTO(S)

ACESSO AO ARTIGO

Documentos Relacionados